TY - JOUR
AU1 - Park, Joong, Yull
AU2 - Takayama,, Shuichi
AU3 - Lee,, Sang-Hoon
AB - Abstract Cells express hundreds of different types of receptors, which they use to continuously monitor their chemical and mechanical microenvironments. Stem cells and cancer cells are particularly sensitive to microenvironmental cues because their interactions have profound effects on stem cell potency and tumorigenesis, respectively. Unlike conventional tissue culture in wells and dishes, microtechnology with dimensions on the cellular scale can be combined with materials, chemicals, physiological flows, and other effectors to provide high levels of control in a format more flexible than macroscale in vitro or in vivo systems, revealing stimulation-specific responses of stem cells and cancer cells. Microtechnology-integrated biology enable the simultaneous control of multiple numbers of biological microenvironmental factors in a high-throughput manner. In this review we present representative examples of the use of microtechnology systems to regulate the mechanical, chemical, topological, adhesive, and other environments of individual stem cells and cancer cells. We then explore the possibilities for simultaneous multimodal control of combinations of these environmental factors. Insight, innovation, integration Microtechnology has proven effective in regulating biological stimuli at the cellular and subcellular scales. Such systems are useful for minimizing confounding effects of the complex in vivo environment to enable deciphering of the role of specific cues and mechanisms on stem cells and cancer cells. Single isolated stimulations such as biochemical and mechanical stimuli have been used in microsystems to clarify their roles in cellular decision-making. Some studies also use multiple combinations of these stimuli to explore combination effects because the combined effects of multiple cues are not always simply additive. Conventional macroscopic tools are likely ineffective to study such a large array of combinations, therefore, this is an opportunity for microtechnology to reveal new cellular behaviors. Introduction Cellsin vivo are exposed, continuously and simultaneously, to different types of biological stimulations, which, in turn, trigger cellular responses important to the cells' biological functions. Stimulation-specific cell responses are particularly important in stem cells and cancer cells, which require a microscopic level of control by biological stimuli to evaluate the exact mechanisms of differentiation, development, tumorigenesis, metastasis, invasion, angiogenesis, and other specific pathways. To achieve these scientifically important objectives, a wide variety of microtechnology-based techniques have been developed to create/control biochemical, mechanical, electrical, surface (topological or surface matrix-related), cellular (cell-to-cell interactions, spheroids), and various other (temperature, magnetic, optical) stimuli (Fig. 1, left). Although cells require specific stimuli to elicit certain responses, they are easily confounded by background or unintentional signals that interfere with the intended signals. Ideally, it should be possible to engineer bioreactors that provide absolute control over the stimulating signals to the cells. However, due to our lack of knowledge of the bandwidths/types of signals that cells respond to and the inherent diversity of cells, observed results to an intended stimulus may include cellular responses from background or collateral stimuli unintentionally created. Thus, continuous efforts have been made to isolate relevant stimuli (Fig. 1, bottom left). Fig. 1 Open in new tabDownload slide Examples of microenvironmental stimuli that regulate cells in microsystems. Isolated stimulation is highly useful for investigating stimulant-specific cellular events (left); however, collateral stimulations can interfere with the desired stimulant-specific cellular responses (bottom left). More physiologic microenvironments can be created by applying multiple stimuli using carefully designed stimulant-inducing mechanisms (right). Multiple stimulations can reveal new features in cell biology by analyzing their superimposed (f(A) + f(B)+…), synergistic, and diverse interacting (f(A,B,…)) effects (bottom right). Fig. 1 Open in new tabDownload slide Examples of microenvironmental stimuli that regulate cells in microsystems. Isolated stimulation is highly useful for investigating stimulant-specific cellular events (left); however, collateral stimulations can interfere with the desired stimulant-specific cellular responses (bottom left). More physiologic microenvironments can be created by applying multiple stimuli using carefully designed stimulant-inducing mechanisms (right). Multiple stimulations can reveal new features in cell biology by analyzing their superimposed (f(A) + f(B)+…), synergistic, and diverse interacting (f(A,B,…)) effects (bottom right). In a trend in microtechnology, researchers have begun to appreciate the importance of controlling multiple stimulations in evaluating higher-order cellular responses in stem cells and cancer cells (Fig. 1, right). For example, mechanical stimuli enhance cell adhesion by triggering the activation of low affinity integrin, which participates in adhesion;1 this type of mechanical stimulus-oriented cascade effect, alone or together with other biochemically-induced stimuli, ultimately results in the differentiation of stem cells.2 Due to the highly organized interlinks of cell receptors that sense various types of stimuli, multiple stimuli can result in additive, antagonistic, synergetic, and other diverse interacting effects on cells (Fig. 1, bottom right). Thus, a deeper understanding of cellular mechanisms requires the regulation and systematic variation of multiple factors. This review will cover the current status of microtechnology-based cell stimulating systems. Various regulatory mechanisms for generating stimulations will be discussed, along with their impact on stem cells and cancer cells. We will also discuss the importance of isolating stimuli and the need to generate multiple stimuli within the microsystems, with examples of biological applications. Stimuli-based stem cell differentiation control in microsystems Future life saving therapies, such as rebuilding the nervous system, repairing damaged heart, curing diabetes, gene therapy, and curing autoimmune diseases, rely on two unique characteristics of stem cells: self-renewal and pluripotency.3 Microscale temporal and spatial regulation of cellsignaling molecules is important for the development and differentiation of stem cells, especially during embryogenesis.4 It is therefore challenging yet critical to identify the mechanisms regulating the differentiation of stem cells. Biochemical stimuli Although cells are able to synthesize, detect, and process biochemical signaling factors to modify their ability to survive under various conditions, the pathways of stem cellsignaling are only partially understood. The use of microsystems has resulted in the accurate discrete localization of specific soluble biomolecules,5 as well as gradual changes of soluble molecule concentrations in microsystems. These molecular patterns and gradients have been shown to be important in stem cell differentiation6 and embryogenesis,7 as well as other biological pathways such as cancer metastasis,8–10wound healing and inflammation,11 and bacterial chemotaxis.12 One implementation of biochemical stimulation in microsystems is in human neural stem cells.13 Following the generation of a tree-type network of planer serpentine microfluidic channels by PDMS (polydimethysiloxane) rapid prototyping and soft lithographic technology,14 concentration gradients of mixtures of growth factors (epidermal growth factor, fibroblastgrowth factor 2, and platelet-derived growth factor) in the cell chamber (cross section of 100 μm × 2.4 mm) were generated and connected to the upper stream serpentine channel network (Fig. 2A). The primary hydrodynamic features of these serpentine microchannels were diffusional mixing and secondary flow mixing, both of which can be predicted with precision.15 The cells were exposed to inducers of proliferation and/or astrocyte differentiation, and differentiation was shown to depend on growth factor concentrations. These studies of biomolecular gradients can be used to generate quantitative profiles of stem cell responses such as differentiation, attachment, and proliferation rates.6,13,16,17 Since stem cells are sensitive to various stimuli, however, a strong convective shear flow may have influenced the results due to the shearing effect of the flow,18 suggesting the need for a clear-cut control experiment to negate unwanted effects such as shear stress. Although it was reported that developmental potential of stem cells are maintained safely under low shear stress (lower than 9.86 dyn/cm2 for mouse embryonic stem cells),19,20 closer investigations should be done further because there are other conflicting reports that describe that even slower flow rates meaningfully altering the cellular behaviors and differentiation significantly in certain cell types.21,22 In addition to that, many microfluidic systems using conventional syringe pumps which easily apply higher shear flow than 10 dyn/cm2 to cells. Therefore, to avoid high shear effects that may interfere with cell response, diffusion-based microsystems23–26 were developed. However, few of these have been applied to stem cell research. Considering the ease of achieving a biomolecular gradient, it is surprising that such gradients have not been widely utilized in studies of stem cells. Long-term stable gradient generation seems to be one of the main obstacles preventing microfluidic systems from being incorporated into stem cell differentiation studies. Thus, more efforts are needed to develop techniques for the generation of long-term (often over weeks) stable concentration gradients of biomolecules. Fig. 2 Open in new tabDownload slide Various stimuli generated in microsystems for stem cell research. (A) Use of a popular serpentine channel network to create a gradient of biomolecules (growth factor) showed that the proliferation/differentiation of hNSCs during 7 days in culture was linearly dependent on the concentration gradient profile. Walls in the channel prevented cell migration. Ref. 13—Reproduced by permission of the Royal Society of Chemistry. (B) Application of pneumatically generated pressure to human mesenchymal stem cells was used to quantify the effect on cell differentiation. Osteogenetic differentiation was found to be enhanced in the pressurized chamber, as shown by immunofluorescent staining with anti CD90 (I—control, II—stimulation) and alkaline phosphatase staining (III—control, IV—stimulation). Ref. 29—Reproduced by permission of the Royal Society of Chemistry. (C) A two-layered microfluidic chip was used to assess the interactions of MDA-MB-231cells (green) and COS7 cells (red) during spheroid formation. PDMS channel layers were separated by a semi-porous polycarbonate membrane on which the cells had been cultured. Cellular patterning could be controlled by geometric features, including shape (diamond, square, bow-tie) and size (not shown). Scale bars: 200 μm. Ref. 49—Reproduced by permission of the Royal Society of Chemistry. (D) Cells were aligned in the direction of the microgrooves on the patterned substrate, but were aligned randomly on the non-patterned substrate. Scale bar: 30 μm. Ref. 50—Reproduced by permission of Elsevier. (E) Protein immobilization vs. microstructures. Quantitative analysis was done for balancing effect of two different stimulations (cytokines and adhesion). At excess cytokine concentrations, adhesion related signaling was overbalanced by an excess of cytokinesignaling, resulting in enhanced stimulation of cell cycling. At moderate cytokine levels, however, increased adhesion inhibited cell cycling. Ref. 51—Reproduced by permission of the Royal Society of Chemistry. (F) Drosophila melanogasterembryo development at two temperatures. The two halves of the embryo, which had been incubated at two different temperatures, were in a different cell cycle, as demonstrated by the difference in nuclear density. Number of nuclei in enlarged areas is shown underneath in yellow. Higher nuclear density was observed in the warmer half of the embryo. Ref. 52—Reproduced by permission of the Nature Publishing Group. Fig. 2 Open in new tabDownload slide Various stimuli generated in microsystems for stem cell research. (A) Use of a popular serpentine channel network to create a gradient of biomolecules (growth factor) showed that the proliferation/differentiation of hNSCs during 7 days in culture was linearly dependent on the concentration gradient profile. Walls in the channel prevented cell migration. Ref. 13—Reproduced by permission of the Royal Society of Chemistry. (B) Application of pneumatically generated pressure to human mesenchymal stem cells was used to quantify the effect on cell differentiation. Osteogenetic differentiation was found to be enhanced in the pressurized chamber, as shown by immunofluorescent staining with anti CD90 (I—control, II—stimulation) and alkaline phosphatase staining (III—control, IV—stimulation). Ref. 29—Reproduced by permission of the Royal Society of Chemistry. (C) A two-layered microfluidic chip was used to assess the interactions of MDA-MB-231cells (green) and COS7 cells (red) during spheroid formation. PDMS channel layers were separated by a semi-porous polycarbonate membrane on which the cells had been cultured. Cellular patterning could be controlled by geometric features, including shape (diamond, square, bow-tie) and size (not shown). Scale bars: 200 μm. Ref. 49—Reproduced by permission of the Royal Society of Chemistry. (D) Cells were aligned in the direction of the microgrooves on the patterned substrate, but were aligned randomly on the non-patterned substrate. Scale bar: 30 μm. Ref. 50—Reproduced by permission of Elsevier. (E) Protein immobilization vs. microstructures. Quantitative analysis was done for balancing effect of two different stimulations (cytokines and adhesion). At excess cytokine concentrations, adhesion related signaling was overbalanced by an excess of cytokinesignaling, resulting in enhanced stimulation of cell cycling. At moderate cytokine levels, however, increased adhesion inhibited cell cycling. Ref. 51—Reproduced by permission of the Royal Society of Chemistry. (F) Drosophila melanogasterembryo development at two temperatures. The two halves of the embryo, which had been incubated at two different temperatures, were in a different cell cycle, as demonstrated by the difference in nuclear density. Number of nuclei in enlarged areas is shown underneath in yellow. Higher nuclear density was observed in the warmer half of the embryo. Ref. 52—Reproduced by permission of the Nature Publishing Group. Mechanical stimulation Mechanical cues are also known to affect stem cell lineages by interacting with biochemical signals in the developmental processes.27 For example, cytoskeletal tension can regulate the lineage of human mesenchymal stem cells (MSCs).28 The major players in mechanical force-induced differentiation include cell-surface stretch receptors, cascades of genes responsible for their synthesis, and the secretion of key ligament extracellular matrix components. Since the osteogenic process of stem cells may depend on mechanical forces, stem cells have been exposed to pressure stimuli in microfluidic system. Pneumatic control of the dynamic compression force was achieved in microfluidic systems and used to investigate the proliferation and osteogenic differentiation of human MSCs (Fig. 2B).29 Application of 1 to 20 kPa mechanical stimulation twice daily for 10 min each over 1, 3, and 7 days was sufficient for stem cell differentiation. ALP (alkaline phosphatase) expression and calcium deposition (a marker of bone matrix deposition) were significantly enhanced in response to pressure, providing quantitative assessment of pressure-sensitive osteogenesis. Other types of mechanical forces, such as strain, can also be effective in promoting osteogenesis. Macroscale experiments were performed using flexible four-point bending substrates to test corresponding bone regeneration.30 This type of system can likely be easily scaled down to take all the advantages of microtechnology, such as high-throughput screening. Due to the major mechanical stresses that can be generated, independent of soluble biochemical stimuli and surface cues, studies using combinations mechanical and other types of stimuli (e.g. shear plus biomolecules, shear plus topology) may be beneficial in elucidating the pathways involved in the development and differentiation of stem cells (Fig. 1, right). Surface stimulation for differentiation of stem cells Surface stimuli differ from other stimuli in that no external actuation or control are necessary to maintain the effect. In terms of mechanotransduction, internal self-generated cell stress caused by surface stimuli31,32 should be carefully compared with stresses caused by direct mechanical stimuli. Topology and extracellular matrix (ECM) are examples of surface stimuli where regulation of physical interactions between cells and the ECM can alter stem cell fate.33 Induction of stem cell differentiation using microcontact printing technology showed that cell behavior was dependent on cell size.34 Various size patterns (approximately 100–500 μm) of alkanethiol (EG4CC1sH) were arrayed on a gold substrate while the other part of the substrate was coated with a hydrophobic alkanethiol (C16SH).35 Human mesenchymal stem cells (hMSCs) were cultured for 10–14 days to quantify adipogenic differentiation. The effect of substrate shape was also discussed; specifically, cells on the periphery of the pattern might sense the edge, eventually affecting the net differentiation of the cells in the interior. The effect of cell–cell interactions on differentiation was not discussed in detail. Ruiz and Chen (2008) used the same microcontact printing technology to pattern human mesenchymal stem cells on PDMS-coated coverslips, focusing not only on the patterned size but also on the stress gradients that preceded and mirrored the differentiation patterns.36 Regions of high stress resulted in osteogenesis while stem cells in regions of low stress differentiated into adipocytes. This study cleverly implemented contact printing technology in research on stem cell differentiation by creating surface stimuli that induced a stress-associated distribution of cells. Nanoscale topographic cues are also important not only for providing structural support for adhesion sites but also triggering topographical signals for cellular differentiation. It has been reported that the concave or convex nano structures might contribute in controlling the closing/opening of ion channels.37 From this, it is postulated that in the context of the in vivo environment, ECM proteins not only communicate through a lock and key mechanism viacell surface receptors, but also serve as nanoscale structures that induce physical deformations in the cell membrane that either directly or indirectly effect cell morphology. Electrospinning is a useful method to create nanofiber structures that resemble morphological ECM collagenfibrils, and was used to confirm the effect collagenfibrils have on selective differentiation of adipogenic, chondrogenic and osteogenic lineages.38 In addition, nanofibers have been coupled with a microfluidic system for hMSC culture to expand versatility of the microtechnology.39 Traditional photoligthographic techniques are still useful in creation of nanoscale geometries in 2D surfaces such as nanograting, nanopost, and nanopit. Using these techniques, fibronectin coated PDMS on which nanostructures (600 nm ridges) were patterned was used to prove that cytoskeletal structures are guided by nanoscale patterns and results in both the morphologic alteration and proliferation of human embryonic stem cells.40 It has to be emphasized that nanotopography is now actively revealing new cellular functions, particularly for stem cell-based tissue engineering, and therefore engineers should focus more on developing and utilizing synthetic nanotopographic substrates.40 The above-mentioned surface type of stimulation can be conveniently combined with other types of stimuli. These combinations can be utilized in multiple stimulation systems required for future microtechnology-based stem cell research. Electrical stimulation Not only is electrical actuation useful for cell manipulation and for sensing cellsignaling, it is also one of the basic stimuli in the body, especially for cardiac cells and nerve cells. Ions such as Ca2+ are ubiquitous in our body to create continuously electrical potential, playing a primary role in maintaining cell life, and even more actively in neural signal transmission. Electrical stimulation has even been appreciated in stem cell research for guiding neural differentiation.41 Endogenous electrical fields (of strength ranging from a few to several hundred mV mm−1) are important in embryonic development.42 For example, disruption of the natural electrical field/gradient in embryos has been found to disrupt development, particularly neural development.42 To date, however, there have been no extensive studies of electrical stimuli in microsystems, although macroscale studies showed that electrical fields promote cardiomyogenesis from embryonic stem cells.43,44 Cellular stimulations in 2D and 3D structures and spheroids Cell-cell contact occurs frequently in cell culture and should be regarded as an important biological stimulus since it changes the structural organization of cells, affecting polarization, migration, and morphogenesis.45 Microtechnology can manipulate small numbers of cellsviacellpatterning/positioning and the formation of two-dimensional (2D) and three-dimensional (3D) cellular structures. For example, a very high-throughput method to pattern murine embryonic stem cells has been designed.46 In this method, a polymeric microchip, called a Bio Flip Chip and containing thousands of microwells that trap single/small numbers of stem cells, showed high patterning efficiency (>75%) on a 2D substrate and another layer of cells. Using this method, cell-cell contact, together with E-cadherin, was shown to repress embryonic stem cell colony formation. By using PDMS microfeatures (microwells, microgrooves), MSC-laden collagen was successfully immobilized in agarose.36 After 14 days in culture, the stem cells showed a partitioned differentiation, with osteogenesis at the periphery of the 3D structure and adipogenesis at its center. This type of partitioned differentiation is thought to be modulated by cytoskeletal tension. Cell-cell adhesion is a selective process (e.g.adipocyte adhesion to β-cells,47 with dysfunction resulting in diabetes) and thus one of the primary events in embryonic development; cells from one tissue specifically adhere to cells of the same tissue rather than to cells of a different tissue.48 This type of preferential cell-cell adhesion is mediated by transmembrane proteins called cell adhesion molecules, including the selectins, the integrins, the immunoglobulin (Ig) superfamily for transient adhesion, and the cadherins for stable adhesion. In addition to these protein-mediated interactions, other cell-cell interaction mechanisms are known, called gap junctions and tight junctions. Microstructures and patterns can be used to regulate these cell-cell interactions and guide differentiation. For example, this type of technology can be used to investigate the development and tissue organization associated with the patterned differentiation of ES cells or embryonic development resulting from inhomogeneous cell-to-cell interactions. For example, the initial position of cells prior to aggregation can influence the final configuration within a co-culture spheroid. Combining cell printing technology with 3D spheroids resulted in a unique method to form Janus 3D spheroids of stem cells49 (Fig. 2C). Multiple stimuli The synergistic effect of mechanical shear with adhesive cell microenvironments (e.g. ECM) has been studied in several systems. For example, the combination of shear effect with the effect of degree of spreading was assessed in osteoblasts. The degree of spreading was controlled by the ECM stamping pattern size (spot diameter, 3 μm; spacing, 5–7 μm), and sinusoidal oscillatory shear flow (1 Pa, 1 Hz) mimicking the dynamic fluid flow in the body, was applied to spreading cells. The shear-related mechanotransduction was shown to change the intercellular calcium balance, resulting in a change in the degree of cell spreading.2 This balance of two different stimuli (focal adhesive cue and fluid shear) revealed that a higher degree of osteoblast spreading was less responsive to shear providing new insight into bone mechanotransduction at the cellular level. The effects of multiple stimuli involving topographical cues on neural progenitor cells was recently assessed, in that three-dimensional topography facilitated neuronal differentiation when combined with substrate adhesive (laminin) and cell adhesive stimuli (Fig. 2D).50 In this study, rat hippocampal progenitor cells co-cultured with astrocytes preferentially acquired neuronal morphology, expressing class III b-tubulin on the micropattern, as opposed to the planar substrate, or compared with hippocampal cells grown in the absence of astrocytes. This study confirmed that integrating multiple cues is important in understanding and controlling neural stem cell differentiation and designing scaffolds for guided nerve regeneration. Another study of multiple stimuli, using surface geometrical cues (microwells) and cytokines,51 mimicked hematopoietic stem cell (HSC) niches in the bone marrow microenvironment. The balance of cytokine level and adhesion spatial restriction was shown to quantitatively affect cell cycling and differentiation (Fig. 2E). This study implies that the combined effect of multiple cues is not always simply additive and thus more systematic comparison studies on the balancing effect of multiple stimuli is necessary. It is these types of parallel studies that will take advantages of the high-throughput potential of microsystems that will be most appreciated. Cells may also be exposed to multiple stimuli of the same type (e.g. multiple biochemical stimuli), especially during early embryonic development, when cells are exposed to various concentrations of many biomolecules. In model systems, these biomolecules have been synthetically applied with high spatiotemporal resolution. A microfluidic system for differentiating neural progenitors by harnessing extremely slow flow driven by an osmotic pump has been proposed.6 Gradients of exogenous cytokines (e.g. sonic hedgehog, bone morphogenetic protein 4) and growth factors (e.g.fibroblastgrowth factor 8), which play central roles in the embryonic development of the neural system, were applied simultaneously in a microfluidic system, resulting in the successful generation of a complex neural network. Opposing effects of agonist (sonic hedgehog) and antagonist (bone morphogenetic protein 4) on the proliferation and differentiation of human ESC-derived neurons was recapitulated quantitatively according to the concentration gradients. It should be noticed that neural system components (brains, notochord, etc) are developed not by simple additive/superposing effect (expressed as ‘f(A) + f(B)+⋯’ in Fig. 2) of each of these biomolecules but by their combinatory effect as multiple stimuli (‘f(A, B, …)’ in Fig. 2). Miscellaneous stimulations The effect of temperature gradient on embryonic differentiation of Drosophila embryo has been investigated using microfluidic system.52 A conventional T-shape microchannel was used to create two discrete phases at different temperatures, the anterior and posterior halves of the embryo developed at different rates, as shown by the density of nuclei in the two halves of the embryo (Fig. 2F). Spatiotemporal control of the environment was successfully achieved in a simple microfluidic system, showing that perturbing the environment is complementary to perturbing the molecular components of the network in the embryo. Other good candidate stimuli for combination with microtechnology are magnetic force and laser irradiation. Magnetic force has been in limited use in microfluidic systems for cell sorting and microbead-related studies. Macroscale studies have shown that magnetic fields inhibit angiogenesis in embryos,53 and that electromagnetic fields affect gene expression of embryonic stem cells54 and mesenchymal stem cells.55,56 Lasers can also be easily combined with microsystems, since the polymers commonly used in these microsystems are transparent. Although lasers have been shown to be osteogenetic stimuli for stem cells, laser stimulation has been infrequently utilized in microfluidics.57 Since microsystems can be inserted easily into systems generating conventional magnetic fields or laser irradiation, these may be promising areas of future research. Microtechnology in cancer cell research In addition to stem cells, cancer cells are also sensitive to microenvironments and thus a very relevant research topic for microtechnology. The most dangerous and unique properties of cancer cells are invasion and metastasis, which result in more lethality than tumor proliferation at primary sites.58 Understanding the mechanism of cancer cell metastasis has been a main focus of cancer research, but this process is not yet fully understood, as shown by the metastatic cascade model.59,60 One of the limiting factors is the lack of in vitrocell culture tools that sufficiently recapitulate the physiological complexities of intravasation, intravascular flow and adhesion, extravasation, and migration to target tissues. Microfluidic technology is a promising tool for recreating aspects of these steps involved in tumor invasion, metastasis, and angiogenesis. Biochemical stimulation One of the most attractive features of microsystems for cancer research is the ability to control chemokine gradients. In a microfluidic chemotaxis experiment using serpentine mixing microchannels, the migration of breast cancer cells was shown to depend on the shape of a gradient of epidermal growth factor (EGF).61 Assessment of subcellular EGFsignaling showed that the ligand-independent propagation of EGF signals occurred for cells with abnormally high receptor density on their plasma membranes, such as carcinoma cells.62 In this system, which utilized 100 μm-wide microchannels with two laminar streams, FRET (Förster resonance energy transfer) indicators of protein phosphorylation and Ras activation were used to assess intracellularsignaling in response to subcellular biochemical stimulation by EGF. Serpentine microchannels were used for another tumor metastasis study to explain the increases in invasive and migratory behavior of cancer cells in close proximity to blood vessels and macrophages, a 1-μm aperture was used to generate a locally intense gradient of biochemicals to observe breast cancer metastasis under such critical conditions.63 Topology and 3D matrix for metastasis Metastasis can also be studied using systems comparable in size to a single cell, since single cells are affected by geometric issues such as blood vessels and fibers.64,65 Although large solid tumors in the body can infuse millions of tumor cells into the blood, only a few cells participate in metastasis via mechanisms that are not fully understood. Metastasis has been modeled using mechanically constrained microchannels, which have features that mimic blood/lymph vessels, collagen fibers, and white matter tracts, and restrict cell migration to one direction.66 Conventional PDMS soft lithographic technology was used to construct straight microchannels, of dimensions 12 × 15 × 600 μm and coated with collagen IV, before seeding cells. The cells were traced for two days in culture. Unlike their motility on flat surfaces, a fast persistent unidirectional migration of individual tumor cells was observed in the absence of external chemical gradients (Fig. 3A). Among the seven cancer cell lines assayed were those derived from lung carcinoma (H446), prostate adenocarcinoma (PC3), and breast adenocarcinoma (MDA-MB 231). Results from these various cell lines showed that their motility phenotype may be dependent on the preferential invasion of different tumor cells along lymph vessels, collagen fibers, and white matter tracts. While this quantitative evaluation of the migration of individual cancer cells is useful clinically, more detailed studies that evaluate the interacting effects of different extracellular matrices or real endothelium/fibers are needed in future. Fig. 3 Open in new tabDownload slide Stimulus generation in microsystems for cancer research. (A) MDA-MB 231 breast cancer cells moved persistently away from the seeding chamber. Geometric constraints (with dimensions on the cellular scale), such as different cross-sectional areas of the channel, affected cell motility, a feature highly important in cancer metastasis. Scale bar: 100 μm. Ref. 66—Reproduced by permission of the Royal Society of Chemistry. (B) A dielectrophoretic (DEP)-microfluidic device, consisting of a PDMS fluidic network (blue), a glass substrate, and patterned ITO electrodes (bottom, scale bar: 100 μm). Shown are a patterned HUVEC-HUVEC pair (right top), a HUVEC-A549 pair (right middle), and a single HUVECcell (right bottom) after 4 h of culture in complete medium. HUVECs are in green and A549 cells are outlined. Cell-cell pairs were generated by the microelectrode system. Ref. 70—Reproduced by permission of the Nature Publishing Group. (C) The physiologic structure of a vessel was mimicked using a three-layered PDMS microfluidic device (top channel, polyester membrane, and bottom channel). Chemokine in the bottom channel activated the endothelium from the basal face, while cancer cells flow in the top channel above the endothelium. Representative fluorescent images of the adhesion of 231-control cells onto region-specific TNF-α treated endothelium under 0.50 dyn/cm2 shear stress flow conditions (middle bottom). In these images, the downstream region was treated with TNF-α for 5 h (denoted by a ‘+’), while the upstream and middle regions were left untreated. Scale bar: 200 μm. Graph: Endothelial cells mediate CXCL12-dependent enhanced adhesion of cancer cells. For all three cancer cell types, adhesion selectivity was towards the CXCL12 treated region of the endothelium over the corresponding untreated region (*). Ref. 76—Reproduced by permission of the author. Fig. 3 Open in new tabDownload slide Stimulus generation in microsystems for cancer research. (A) MDA-MB 231 breast cancer cells moved persistently away from the seeding chamber. Geometric constraints (with dimensions on the cellular scale), such as different cross-sectional areas of the channel, affected cell motility, a feature highly important in cancer metastasis. Scale bar: 100 μm. Ref. 66—Reproduced by permission of the Royal Society of Chemistry. (B) A dielectrophoretic (DEP)-microfluidic device, consisting of a PDMS fluidic network (blue), a glass substrate, and patterned ITO electrodes (bottom, scale bar: 100 μm). Shown are a patterned HUVEC-HUVEC pair (right top), a HUVEC-A549 pair (right middle), and a single HUVECcell (right bottom) after 4 h of culture in complete medium. HUVECs are in green and A549 cells are outlined. Cell-cell pairs were generated by the microelectrode system. Ref. 70—Reproduced by permission of the Nature Publishing Group. (C) The physiologic structure of a vessel was mimicked using a three-layered PDMS microfluidic device (top channel, polyester membrane, and bottom channel). Chemokine in the bottom channel activated the endothelium from the basal face, while cancer cells flow in the top channel above the endothelium. Representative fluorescent images of the adhesion of 231-control cells onto region-specific TNF-α treated endothelium under 0.50 dyn/cm2 shear stress flow conditions (middle bottom). In these images, the downstream region was treated with TNF-α for 5 h (denoted by a ‘+’), while the upstream and middle regions were left untreated. Scale bar: 200 μm. Graph: Endothelial cells mediate CXCL12-dependent enhanced adhesion of cancer cells. For all three cancer cell types, adhesion selectivity was towards the CXCL12 treated region of the endothelium over the corresponding untreated region (*). Ref. 76—Reproduced by permission of the author. Three-dimensional matrices require in depth analysis using microtechnology. Cancer cell migration and invasion are dependent on the 3D matrix structure around the cells, due to the interplay and remodeling of ECM microstructures.67 3D migration of cancer cells differs from 2D migration because the former involves a more active and complex interaction between cells and the 3D matrix. A controlled/aligned matrix is thus highly desirable for quantitative assessments of cancer cell metastasis in vivo, and 3D polymerization of collagen type I, an ECM component, was successfully achieved in an array-based high-throughput microsystem;68 allowing a careful investigation of microscale collagenpolymerization and of cellular reaction to microcollagen fiber thickness. Due to the importance of collagen in tumor research, including the similar diffusivity of biomolecules through collagen fibers and tumor tissue,69 microsystems involving collagen or other 3D ECM matrices may be very helpful in research on cancer metastasis/invasion and in the treatment of cancer patients. Cellular stimulations for cancer research As mentioned in the section on stem cells, intercellular communication is essential for cell development and the function of most cell types, including cancer cells. One of the most important factors involved in tumor growth, invasion and metastasis is blood vessel formation in or around tumor tissue. One-to-one cell pairing may be the most appropriate method to closely examine highly organized cell-to-cell interactions that occur during tumor microvascularization. A microfluidic device harnessing dielectrophoretic electrodes has been utilized to manipulate cells precisely (Fig. 3B).70 Pairing endothelial cells (EC; e.g.HUVEC) and endothelial tumor cells (e.g. A549, a human lung cancer cell line) showed that migration speed and migration patterns were strongly affected by secreted collagen and VEGF. Inhibition of VEGF-mediated receptoractivation did not significantly affect the movement of these A549–HUVEC pairs, but, surprisingly, had a dramatic effect on the movement of single HUVECs and HUVEC–HUVEC pairs. Interestingly, VEGF secretion by cancer cells did not cause endothelial cell migration, indicating that more complex parameters are involved in tumor angiogenesis. This study, however, directly confirmed that cellular stimulation plays an important role in cancer growth and metastasis. The induction and restriction of direct cell-cell contact was conveniently achieved in a micropatterned substrate,71 with the latter dependent on the size of the adhesive patterns. One to a few cells were collected at each spot on the array, with clusters of tumor cells showing enhanced proliferative and angiogenic potential compared with individual cells. Electrical stimulation Transformed (cancerous) cells may be subject to electrical perturbation, as they have a greater negative surface charge than normal cells.72 In addition, subjection of cancer cells and neighboring normal cells to an extracellular electrical field gradient resulted in the preferential migration of cancer cells.73 This electrotaxis was reproduced in a multiple-electric-field microsystem using cultured lung cancer cells.74 Embryonic carcinoma cells and embryonic stem cells also have been shown to reprogram somatic cells, with fusion being the crucial step. A microfluidic system was used to investigate cell fusion induced by electrical stimulation.75 Using double-layered pocket-shaped microstructures, fluid dynamics, and microelectrode patterning, a highly efficient high-throughput cell-pairing/fusion was achieved, showing that electrical stimulation can be a good inducer of unique cell–cell interactions of cells. This may allow studies of cancer cell reprogramming processes as well as being used as a direct stimulator of cancer cells. Physiological model using microfluidic system Since static in vitro assays of cancer cells may lead to results that are not relevant in vivo, a more physiologic assay was designed, which used a microfluidic system and breast cancer cells that interact with endothelium (Fig. 3C).76 This system, which closely mimics the physiologic conditions of circulating cancer cells that adhere to endothelium in organs, consists of two PDMS layers sandwiching a thin porous polyester membrane on which a confluent monolayer of human dermal microvascular endothelial cells (HDMECs) are cultured. Region-specific stimulation of endothelium with chemokines enabled a comparison of the adhesion to endothelium of cancer cells of differing metastatic potential. This system showed that CXCL12 acts through receptor CXCR4 on endothelium to promote adhesion of circulating breast cancer cells. Multiple stimulations for cancer research Stimulation of cancer cells with multiple stimuli can result in new features, which were not observed when these stimuli were applied separately. For example, although a gradient of CXCL12, a ligand involved in various types of cancer cells including breast cancer cells, did not cause chemotaxis, this CXCL12 gradient, together with a uniform stimulus with EGF, resulted in directional cancer cell migration.9 This result indicates that metastatic cells are under the control of multiple stimuli, such as superimposed gradients of growth factors and chemokines. Studies using multiple stimuli are also useful for expanding the results obtained from studies of a single stimulus. For example, the mechanisms responsible for cancer cell migration were studied by assessing the superimposed effects of drugs (nocodazole and paclitaxel) on topological confinement (microchannel 12 × 15 μm).66 Geometric cues by themselves stimulate cancer cells to move persistent and fast in one direction. When drug was added to this system, although the average cancer cell migration speed dropped, a small number of cancer cells still migrated as if not affected by the drug, enabling the identification of unique subpopulations of cancer cells based on mobility in confined structural microenvironments and microtubule disrupting drugs. Screening such cancer cell subpopulations may be useful for cancer metastasis research. Regulation of microenvironmental stimulation Isolation of a single stimulant Studies that provide a first-order cellular response to a single specific stimulant would enable the most straightforward mechanistic interpretations. This is rarely the case, inasmuch as various environmental cues act simultaneously as cell stimulants. For example, cells may be stimulated by substrate stiffness, ECM, cell-cell interactions, and nutrient concentrations, as well as by specific stimuli. In addition, some mechanisms used in microsystems to generate a certain type of stimulation can cause collateral stimulations. One representative example is gradient-generation microfluidic systems, which have been used, for example, in studies of embryo development52 or to stimulate other cell types.5,77 This type of microfluidic system basically relies on laminar streams to create gradients. However, the gradients used in these studies are generally super-physiological in terms of steepness. A critical study quantified the effects of flow, not the gradient, on cell migration in response to the chemokine CXCL878 (Fig. 4A). Therefore it should be emphasized that biomolecular studies in microfluidic devices can be significantly altered by the effects of flow. Other possible examples are topological cues that can affect local fluid dynamics and concentration-dependent distributions of biomolecules. Designing a microsystem that can isolate an individual stimulus while minimizing collateral stimuli will be most desirable; for example, a ladder type microchamber device has been developed that achieved a shear-free gradient for long-term use (Fig. 4B).26 Alternatively, well-defined control tests should be performed to prevent these unwanted effects from interfering with analysis of the data. Fig. 4 Open in new tabDownload slide Detection, prevention and creation of collateral/multiple stimulations. (A) Collateral shear stimulation of cells caused by a convective flow generating a concentration gradient in a microfluidic system. A serpentine network was used to create a concentration gradient and the movement of each cell was analyzed by measuring its cumulative distance travelled along perpendicular axes (x and y) (left). Representative results show cell migration trajectories in a chemokine gradient with a flow rate of 3 μL min−1 over 20 min (right). Higher flow rates (6 and 20 μL min−1), however, guided the cells down the microchannels in the direction of flow, thus biasing their chemotactic response. Ref. 78—Reproduced by permission of the Royal Society of Chemistry. (B) Use of a ladder-type microchannel to prevent shear effects. Unlike the conventional gradient generation by two-phase flows, the rung part of the ladder channel was exposed only to diffusion. Fluorescent images show that a concentration gradient could be easily achieved by diffusion alone (top: 2D chamber, middle: 3D Matrigel, bottom: 3D Matrigel with different gradient profile). The experimental gradient profiles (black) showed a good match to their theoretical counterparts (gray) (intensity profiles taken across the white dashed line). Ref. 26—Reproduced by permission of the American Chemical Society. (C) The circular geometry of the microchannels was useful in create gradients of multiple stimulants (shear and concentration). Diffusion strength of biomolecules was carefully considered to decision on dimensions of channel. Shear stress distribution was reasonably wide; about 4-fold difference between minimum and maximum value was achieved. Ref. 22—Reproduced by permission of the Royal Society of Chemistry. Fig. 4 Open in new tabDownload slide Detection, prevention and creation of collateral/multiple stimulations. (A) Collateral shear stimulation of cells caused by a convective flow generating a concentration gradient in a microfluidic system. A serpentine network was used to create a concentration gradient and the movement of each cell was analyzed by measuring its cumulative distance travelled along perpendicular axes (x and y) (left). Representative results show cell migration trajectories in a chemokine gradient with a flow rate of 3 μL min−1 over 20 min (right). Higher flow rates (6 and 20 μL min−1), however, guided the cells down the microchannels in the direction of flow, thus biasing their chemotactic response. Ref. 78—Reproduced by permission of the Royal Society of Chemistry. (B) Use of a ladder-type microchannel to prevent shear effects. Unlike the conventional gradient generation by two-phase flows, the rung part of the ladder channel was exposed only to diffusion. Fluorescent images show that a concentration gradient could be easily achieved by diffusion alone (top: 2D chamber, middle: 3D Matrigel, bottom: 3D Matrigel with different gradient profile). The experimental gradient profiles (black) showed a good match to their theoretical counterparts (gray) (intensity profiles taken across the white dashed line). Ref. 26—Reproduced by permission of the American Chemical Society. (C) The circular geometry of the microchannels was useful in create gradients of multiple stimulants (shear and concentration). Diffusion strength of biomolecules was carefully considered to decision on dimensions of channel. Shear stress distribution was reasonably wide; about 4-fold difference between minimum and maximum value was achieved. Ref. 22—Reproduced by permission of the Royal Society of Chemistry. Regulation of multiple stimulations The meaning and concept of multiple stimuli (or combined stimuli) should be carefully considered. In this paper, the term ‘multiple’ was applied to systems in which stimulation by more than one agent was designed by the researchers for strong biological reasons. These stimuli differ from unwanted side effects or regular background stimuli, such as ECM effect in shear experiments and shear effect in biomolecular experiments. The regulation of multiple stimulations is a promising strategy for future biological integrated microtechnologies. It should be noted that the response of a cell to a single stimulation can differ, depending on if and when a cell is exposed to other stimuli. Moreover, many types of cellsin vivo are constantly exposed to various forces simultaneously. Some cell responses may be due to a superimposition of several individual responses (Fig. 1, bottom right, f(A) + f(B)), whereas others may be synergistic or interfering (f(A)f(B)). For example, growth factors are conventionally used to grow cells and increase attachment and proliferation rate,16 and thus, its effect is sometimes just additive (f(A) + f(B)) with other stimulations such as shear stress.22 However, a certain type of growth factor such as fibroblastgrowth factor 8 has a strong regulatory function when it is applied to stem cells with different stimulus agents such as sonic hedgehog and bone morphogenetic protein 4; most of neuronal systems including fore/mid/hind brains, neural tube, and notochord are the results of the combination of these three key cues.6 Therefore this type of synergistic/interacting effect (f(A, B, …)) of multi-stimuli is not able to be derived from experiments that analyze these cues separately. The mechanism underlying cancer cell migration/metastasis66 and precise guidelines regarding stem cell differentiation should be extensively studied with combinations of two or more multiple stimuli. Thus, there is a need to regulate and systematically vary multiple factors to evaluate their additive, synergistic, antagonistic, or non-interacting effects. Scientific findings on the balance between two or more stimuli require well-defined categorization and appropriate biologic backgrounds, as well as comparisons resulting from varying the amount of stimulant(s) and the time period of cell exposure to them (Fig. 2E).51 Although avoiding unwanted collateral stimuli may be challenging, so may deliberately combining different types of desired stimuli, particularly when combining fluid shear stimulation with biochemical stimulation.78 One strategy to simultaneously regulate biochemical and mechanical shear stimulation may result from using the microchannel geometry. For example, a shear flow gradient may be generated in a circular channel, providing a faster flow to the inner side, with a biomolecular concentration gradient superimposed on this flow by natural diffusion (Fig. 4C).22 This type of device provides several advantages, including the ability to assess various cellular responses in a single experiment, such that cell adhesion, mobility, and attachment time (which may be strongly affected by shear flow even at an interstitial level), and cell proliferation (which is dependent on nutrient level). Meanwhile, most surface stimuli (by control of topology and adhesive cell microenvironments)36,50,51 and cellular stimuli (by spheroids or micro cell-patterning)49 can be easily combined with other types of stimuli such as mechanical shear or strain stimuli and biochemical soluble stimulants. For example, using a recently reported aqueous cell printing methods79 can be conveniently combined with stretchable substrates (strain stimulus) or microfluidic channels (shear stimulus) which is possibly useful for revealing osteogenetic mechanisms in detail. In addition, most of the currently available microfluidic cell culture systems can be implemented with additional electrical and magnetic fields, and perturbing temperature stimulation,52 which might provide newer visions for molecular components interplay in embryonic development. It is this type of multi-stimuli-involved researches that will be most important in finding the balance between the candidate stimulations that is also important because not only the existence but also the portion of each stimulant is critical. In this regard, it enables microtechnologies to capitalize on their unprecedented advantages in parallel high-throughput abilities, for example, in culturing cells,80 merging cells within biochemical capsuls,81cell entrapment and handling,75etc. Conclusion and outlook Microtechnology has contributed greatly to cellular biology in recent decades, most notably in research on stem cells and cancer cells. Although the stimulation-specific responses of these important cells have been extensively studied in microsystems, further improvements in these stimulus-generating microsystems are necessary to reveal high-order biological mechanisms. The control of multiple stimuli in an in vitro microsystem is an important advance beyond single stimulation systems, since multi-stimulation is a physiological norm. Various new cell behaviors have been identified using different stimulants simultaneously, behaviors that may not have been derived from single-stimulation systems. Thus, these new types of microsystems will be at the forefront of stem cell and cancer research in the future. Additionally, coupling the multiple stimuli microsystem design concept with multiple microscale analysis, including the detection and analysis of optical, electrical, mechanical, and single molecule stimuli, offers exciting potential for future applications in cellular biology. For the biologists, the opportunity is to identify existing tools and combinations of tools to reveal new cellular behaviors. For the technologists, the opportunity is to develop new tools that are user-friendly, scalable for high-throughput studies, and versatile for combining various cell stimulation tools. Acknowledgements J. Y. Park was supported by the Korea Research Foundation Grant, Republic of Korea (KRF-2008-357-D00030). This study was supported by a grant of the NRL (National Research Lab) program, the Korea Science and Engineering Foundation (KOSEF), Republic of Korea (20090083115). S. Takayama also thanks the NIH (R01 CA136829-01; R01 HL084370-04; R01 HG004653-02) for support. References 1 M. A. Schwartz and D. W. DeSimone, Curr. Opin. Cell Biol. , 2008 , 20 , 551 – 556 . Crossref Search ADS PubMed 2 L. You , S. Temiyasathit, S. R. Coyer, A. J. Garcia and C. R. Jacobs, Cell. Mol. Bioeng. , 2008 , 1 , 182 – 188 . Crossref Search ADS 3 R. Kirschstein and L. R. Skirboll, Report Prepared by The National Institutes of Health , 2001 , 46 . 4 P. P. Tam and D. A. Loebel, Nat. Rev. Genet. , 2007 , 8 , 368 – 381 . Crossref Search ADS PubMed 5 S. Takayama , E. Ostuni, P. LeDuc, K. Naruse, D. E. Ingber and G. M. Whitesides, Nature , 2001 , 411 , 1016 . Crossref Search ADS PubMed 6 J. Y. Park , S. K. Kim, D. H. Woo, E. J. Lee, J. H. Kim and S. H. Lee, Stem Cells , 2009 . 7 T. N. Behar , A. E. Schaffner, C. A. Colton, R. Somogyi, Z. Olah, C. Lehel and J. L. Barker, J. Neurosci. , 1994 , 14 , 29 – 38 . Crossref Search ADS PubMed 8 A. Muller , B. Homey, H. Soto, N. Ge, D. Catron, M. E. Buchanan, T. McClanahan, E. Murphy, W. Yuan, S. N. Wagner, J. L. Barrera, A. Mohar, E. Verastegui and A. Zlotnik, Nature , 2001 , 410 , 50 – 56 . Crossref Search ADS PubMed 9 B. Mosadegh , W. Saadi, S. J. Wang and N. L. Jeon, Biotechnol. Bioeng. , 2008 , 100 , 1205 – 1213 . Crossref Search ADS PubMed 10 W. Saadi , S. J. Wang, F. Lin and N. L. Jeon, Biomed. Microdevices , 2006 , 8 , 109 – 118 . Crossref Search ADS PubMed 11 P. C. Wilkinson , Chemotaxis and Inflammation , Churchill Livingstone , Edinburgh, New York , 1982 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 12 J. P. Diao , L. Young, S. Kim, E. A. Fogarty, S. M. Heilman, P. Zhou, M. L. Shuler, M. M. Wu and M. P. DeLisa, Lab Chip , 2006 , 6 , 381 – 388 . Crossref Search ADS PubMed 13 B. G. Chung , L. A. Flanagan, S. W. Rhee, P. H. Schwartz, A. P. Lee, E. S. Monuki and N. L. Jeon, Lab Chip , 2005 , 5 , 401 – 406 . Crossref Search ADS PubMed 14 Y. N. Xia and G. M. Whitesides, Angew. Chem. Int. Edit. , 1998 , 37 , 551 – 575 . Crossref Search ADS 15 J. T. Yang and K. W. Lin, J. Micromech. Microeng. , 2006 , 16 , 2439 – 2448 . Crossref Search ADS 16 J. Y. Park , C. M. Hwang and S. H. Lee, Lab Chip , 2007 , 7 , 1673 – 1680 . Crossref Search ADS PubMed 17 T. M. Keenan and A. Folch, Lab Chip , 2008 , 8 , 34 – 57 . Crossref Search ADS PubMed 18 E. Figallo , C. Cannizzaro, S. Gerecht, J. A. Burdick, R. Langer, N. Elvassore and G. Vunjak-Novakovic, Lab Chip , 2007 , 7 , 710 – 719 . Crossref Search ADS PubMed 19 L. Kim , M. D. Vahey, H. Y. Lee and J. Voldman, Lab Chip , 2006 , 6 , 394 – 406 . Crossref Search ADS PubMed 20 E. Y. Fok and P. W. Zandstra, Stem Cells , 2005 , 23 , 1333 – 1342 . Crossref Search ADS PubMed 21 J. P. Stegemann , H. Hong and R. M. Nerem, J. Appl. Physiol. , 2005 , 98 , 2321 – 2327 . Crossref Search ADS PubMed 22 J. Y. Park , S. J. Yoo, C. M. Hwang and S. H. Lee, Lab Chip , 2009 , 9 , 2194 – 2202 . Crossref Search ADS PubMed 23 B. G. Chung , F. Lin and N. L. Jeon, Lab Chip , 2006 , 6 , 764 – 768 . Crossref Search ADS PubMed 24 V. V. Abhyankar , M. A. Lokuta, A. Huttenlocher and D. J. Beebe, Lab Chip , 2006 , 6 , 389 – 393 . Crossref Search ADS PubMed 25 W. Saadi , S. W. Rhee, F. Lin, B. Vahidi, B. G. Chung and N. L. Jeon, Biomed. Microdevices , 2007 , 9 , 627 – 635 . Crossref Search ADS PubMed 26 B. Mosadegh , C. Huang, J. W. Park, H. S. Shin, B. G. Chung, S. K. Hwang, K. H. Lee, H. J. Kim, J. Brody and N. L. Jeon, Langmuir , 2007 , 23 , 10910 – 10912 . Crossref Search ADS PubMed 27 K. Bhadriraju , M. Yang, S. Alom Ruiz, D. Pirone, J. Tan and C. S. Chen, Exp. Cell Res. , 2007 , 313 , 3616 – 3623 . Crossref Search ADS PubMed 28 R. McBeath , D. M. Pirone, C. M. Nelson, K. Bhadriraju and C. S. Chen, Dev. Cell , 2004 , 6 , 483 – 495 . Crossref Search ADS PubMed 29 W. Y. Sim , S. W. Park, S. H. Park, B. H. Min, S. R. Park and S. S. Yang, Lab Chip , 2007 , 7 , 1775 – 1782 . Crossref Search ADS PubMed 30 M. C. Qi , J. Hu, S. J. Zou, H. Q. Chen, H. X. Zhou and L. C. Han, Int. J. Oral Maxillofacial Surg. , 2008 , 37 , 453 – 458 . Crossref Search ADS 31 C. M. Nelson , R. P. Jean, J. L. Tan, W. F. Liu, N. J. Sniadecki, A. A. Spector and C. S. Chen, Proc. Natl. Acad. Sci. U. S. A. , 2005 , 102 , 11594 – 11599 . Crossref Search ADS PubMed 32 D. E. Ingber , Proc. Natl. Acad. Sci. U. S. A. , 2005 , 102 , 11571 – 11572 . Crossref Search ADS PubMed 33 F. Guilak , D. M. Cohen, B. T. Estes, J. M. Gimble, W. Liedtke and C. S. Chen, Cell Stem Cell , 2009 , 5 , 17 – 26 . Crossref Search ADS PubMed 34 C. S. Chen , M. Mrksich, S. Huang, G. M. Whitesides and D. E. Ingber, Science , 1997 , 276 , 1425 – 1428 . Crossref Search ADS PubMed 35 W. Luo , S. R. Jones and M. N. Yousaf, Langmuir , 2008 , 24 , 12129 – 12133 . Crossref Search ADS PubMed 36 S. A. Ruiz and C. S. Chen, Stem Cells , 2008 , 26 , 2921 – 2927 . Crossref Search ADS PubMed 37 V. Vogel and M. Sheetz, Nat. Rev. Mol. Cell Biol. , 2006 , 7 , 265 – 275 . Crossref Search ADS PubMed 38 W. J. Li , R. Tuli, X. Huang, P. Laquerriere and R. S. Tuan, Biomaterials , 2005 , 26 , 5158 – 5166 . Crossref Search ADS PubMed 39 K. H. Lee , G. H. Kwon, S. J. Shin, J. Y. Baek, D. K. Han, Y. Park and S. H. Lee, J. Biomed. Mater. Res., Part A , 2009 , 90A , 619 – 628 . Crossref Search ADS 40 C. J. Bettinger , R. Langer and J. T. Borenstein, Angew. Chem., Int. Ed. , 2009 , 48 , 5406 – 5415 . Crossref Search ADS 41 M. Yamada , K. Tanemura, S. Okada, A. Iwanami, M. Nakamura, H. Mizuno, M. Ozawa, R. Ohyama-Goto, N. Kitamura, M. Kawano, K. Tan-Takeuchi, C. Ohtsuka, A. Miyawaki, A. Takashima, M. Ogawa, Y. Toyama, H. Okano and T. Kondo, Stem Cells , 2007 , 25 , 562 – 570 . Crossref Search ADS PubMed 42 C. D. McCaig , A. M. Rajnicek, B. Song and M. Zhao, Physiol. Rev. , 2005 , 85 , 943 – 978 . Crossref Search ADS PubMed 43 E. Serena , E. Figallo, N. Tandon, C. Cannizzaro, S. Gerecht, N. Elvassore and G. Vunjak-Novakovic, Exp. Cell Res. , 2009 . 44 Y. Kawahara , K. Yamaoka, M. Iwata, M. Fujimura, T. Kajiume, T. Magaki, M. Takeda, T. Ide, K. Kataoka, M. Asashima and L. Yuge, Pathobiology , 2006 , 73 , 288 – 294 . Crossref Search ADS PubMed 45 R. A. Desai , L. Gao, S. Raghavan, W. F. Liu and C. S. Chen, J. Cell Sci. , 2009 , 122 , 905 – 911 . Crossref Search ADS PubMed 46 A. Rosenthal , A. Macdonald and J. Voldman, Biomaterials , 2007 , 28 , 3208 – 3216 . Crossref Search ADS PubMed 47 S. Santeesson , E. Dergerman, P. Rorsman, T. Johansson, S. Lemos and S. Nilsson, Integr. Biol. , 2009 , 1 , 595 – 601 . Crossref Search ADS 48 G. M. Cooper , and R. E. Hausman, The Cell: A Molecular Approach , ASM Press, Sinauer Associates , Washington, DC, Sunderland, Mass , 2007 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 49 Y. Torisawa , B. Mosadegh, G. D. Luker, M. Morell, K. S. O'shea and S. Takayama, Intergative Biology , 2009 . 50 J. B. Recknor , D. S. Sakaguchi and S. K. Mallapragada, Biomaterials , 2006 , 27 , 4098 – 4108 . Crossref Search ADS PubMed 51 I. Kurth , K. Franke, T. Pompe, M. Bornhauser and C. Werner, Integr. Biol. , 2009 , 1 , 427 – 434 . Crossref Search ADS 52 E. M. Lucchetta , J. H. Lee, L. A. Fu, N. H. Patel and R. F. Ismagilov, Nature , 2005 , 434 , 1134 – 1138 . Crossref Search ADS PubMed 53 M. Ruggiero , D. P. Bottaro, G. Liguri, M. Gulisano, B. Peruzzi and S. Pacini, Bioelectromagnetics , 2004 , 25 , 390 – 396 . Crossref Search ADS PubMed 54 J. Czyz , K. M. Guan, Q. H. Zeng, T. Nikolova, A. Meister, F. Schonborn, J. Schuderer, N. Kuster and A. M. Wobus, Bioelectromagnetics , 2004 , 25 , 296 – 307 . Crossref Search ADS PubMed 55 L. Y. Sun , D. K. Hsieh, P. C. Lin, H. T. Chiu and T. W. Chiou, Bioelectromagnetics , 2009 . 56 Z. Wang , P. Yang, H. Xu, A. Qian, L. Hu and P. Shang, Bioelectromagnetics , 2009 , 30 , 446 – 453 . Crossref Search ADS PubMed 57 H. K. Kim , J. H. Kim, A. A. Abbas, D. O. Kim, S. J. Park, J. Y. Chung, E. K. Song and T. R. Yoon, Lasers Med. Sci. , 2009 , 24 , 214 – 222 . Crossref Search ADS PubMed 58 M. B. Sporn , Lancet , 1996 , 347 , 1377 – 1381 . Crossref Search ADS PubMed 59 K. Podsypanina , Y. C. N. Du, M. Jechlinger, L. J. Beverly, D. Hambardzumyan and H. Varmus, Science , 2008 , 321 , 1841 – 1844 . Crossref Search ADS PubMed 60 C. A. Klein , Science , 2008 , 321 , 1785 – 1787 . Crossref Search ADS PubMed 61 S. J. Wang , W. Saadi, F. Lin, C. Minh-Canh Nguyen and N. Li Jeon, Exp. Cell Res. , 2004 , 300 , 180 – 189 . Crossref Search ADS PubMed 62 A. Sawano , S. Takayama, M. Matsuda and A. Miyawaki, Dev. Cell , 2002 , 3 , 245 – 257 . Crossref Search ADS PubMed 63 S. Fok , P. Domachuk, G. Rosengarten, N. Krause, F. Braet, B. J. Eggleton and L. L. Soon, Biophys. J. , 2008 , 95 , 1523 – 1530 . Crossref Search ADS PubMed 64 H. Gerhardt and H. Semb, J. Mol. Med. , 2008 , 86 , 135 – 144 . Crossref Search ADS PubMed 65 E. Sahai , Nat. Rev. Cancer , 2007 , 7 , 737 – 749 . Crossref Search ADS PubMed 66 D. Irimia and M. Toner, Integr. Biol. , 2009 , 1 , 506 – 512 . Crossref Search ADS 67 P. P. Provenzano , D. R. Inman, K. W. Eliceiri, S. M. Trier and P. J. Keely, Biophys. J. , 2008 , 95 , 5374 – 5384 . Crossref Search ADS PubMed 68 K. E. Sung , G. Su, C. Pehlke, S. M. Trier, K. W. Eliceiri, P. J. Keely, A. Friedl and D. J. Beebe, Biomaterials , 2009 , 30 , 4833 – 4841 . Crossref Search ADS PubMed 69 S. Ramanujan , A. Pluen, T. D. McKee, E. B. Brown, Y. Boucher and R. K. Jain, Biophys. J. , 2002 , 83 , 1650 – 1660 . Crossref Search ADS PubMed 70 Z. Yin , D. Noren, C. J. Wang, R. Hang and A. Levchenko, Mol. Syst. Biol. , 2008 , 4 , 232 . Crossref Search ADS PubMed 71 C. P. Tan , B. R. Seo, D. J. Brooks, E. M. Chandler, H. G. Craighead and C. Fischbach, Integr. Biol. , 2009 , 1 , 587 – 594 . Crossref Search ADS 72 E. J. Ambrose , A. M. James and J. H. B. Lowick, Nature , 1956 , 177 , 576 – 577 . Crossref Search ADS PubMed 73 M. E. Mycielska and M. B. A. Djamgoz, J. Cell Sci. , 2004 , 117 , 1631 – 1639 . Crossref Search ADS PubMed 74 C. W. Huang , J. Y. Cheng, M. H. Yen and T. H. Young, Biosens. Bioelectron. , 2009 , 24 , 3510 – 3516 . Crossref Search ADS PubMed 75 A. M. Skelley , O. Kirak, H. Suh, R. Jaenisch and J. Voldman, Nat. Methods , 2009 , 6 , 147 – 152 . Crossref Search ADS PubMed 76 J. W. Song , S. P. Cavnar, A. C. Walker, K. E. Luker, M. Gupta, Y. C. Tung, G. D. Luker and S. Takayama, PLoS One , 2009 , 4 , e5756 . Crossref Search ADS PubMed 77 N. Li Jeon , H. Baskaran, S. K. Dertinger, G. M. Whitesides, L. Van de Water and M. Toner, Nat. Biotechnol. , 2002 , 20 , 826 – 830 . Crossref Search ADS PubMed 78 G. M. Walker , J. Sai, A. Richmond, M. Stremler, C. Y. Chung and J. P. Wikswo, Lab Chip , 2005 , 5 , 611 – 618 . Crossref Search ADS PubMed 79 H. Tavana , A. Jovic, B. Mosadegh, Q. Y. Lee, X. Liu, K. E. Luker, G. D. Luker, S. J. Weiss and S. Takayama, Nat. Mater. , 2009 , 8 , 736 – 741 . Crossref Search ADS PubMed 80 P. J. Hung , P. J. Lee, P. Sabounchi, R. Lin and L. P. Lee, Biotechnol. Bioeng. , 2005 , 89 , 1 – 8 . Crossref Search ADS PubMed 81 M. Chabert and J. L. Viovy, Proc. Natl. Acad. Sci. U. S. A. , 2008 , 105 , 3191 – 3196 . Crossref Search ADS PubMed Joong Yull Park Open in new tabDownload slide Open in new tabDownload slide Joong Yull Park is a post-doctoral fellow in the Department of Biomedical Engineering at the University of Michigan, Ann Arbor since 2009. He received his BS and MS degrees in mechanical engineering from the Chung-Ang University and KAIST, respectively, and his PhD in biomedical engineering in Seoul National University in 2006. He worked as a research professor at Korea University in 2008. His research interests are microfluidics, biorheology, and characterization of cell mechanics. Shuichi Takayama Open in new tabDownload slide Open in new tabDownload slide Shuichi Takayama is Associate Professor in the Department of Biomedical Engineering and the Macromolecular Science and Engineering Center at the University of Michigan, Ann Arbor. He received his BS and MS from the University of Tokyo in 1994 and his PhD degree in chemistry from the Scripps Research Institute in 1998, after which he did postdoctoral studies at Harvard University as a Leukemia and Lymphoma Society postdoctoral fellow. He joined the faculty of the Department of Biomedical Engineering at the University of Michigan, Ann Arbor, in the fall of 2000. His research focuses on engineering cellular microenvironments. Sang-Hoon Lee Open in new tabDownload slide Open in new tabDownload slide Sang-Hoon Lee is professor in the department of Biomedical Engineering at Korea University. He received the BS degree in electrical engineering and the MS and PhD degrees in biomedical engineering from the Seoul National University in Korea, 1983, 1987 and 1992, respectively. From 1992 to 2006, he was Professor in the Department of Biomedical Engineering at the Dankook University. His current interests are the development of microfluidic devices to provide microenvironment for cell study and tissue engineering, microfluidic fiber and capsule fabrication for tissue engineering, and flexible implantable sensor for biomedical applications. This journal is © The Royal Society of Chemistry 2010
TI - Regulating microenvironmental stimuli for stem cells and cancer cells using microsystems
JF - Integrative Biology
DO - 10.1039/c000442a
DA - 2010-06-08
UR - https://www.deepdyve.com/lp/oxford-university-press/regulating-microenvironmental-stimuli-for-stem-cells-and-cancer-cells-ayxhWUrGw0
SP - 229
EP - 240
VL - 2
IS - 5-6
DP - DeepDyve
ER -